Enhanced Uniqueness for Pattern Recognition

1. A method comprising: storing a reference image of a test structure, said test structure comprising a first set of features and a second set of features, said first set of features comprising a subset that is extracted from product features to be monitored in a chip, said second set of features comprising a rotating, space scaling, and linewidth scaling of said first set of features, said second set disposed adjacent to said first set, said second set representing 3 to 15 percent of said first set, said second set being distinguishable from other similar structures in the vicinity to provide uniqueness to said test structure; loading a sample on a stage; moving said stage; focusing and fine-tuning a test image of said test structure on said sample; capturing said test image; scanning a plurality of portions of said test image; performing pattern recognition of each of said portions; evaluating similarity of said test image to said reference image; determining a score for each of said portions; ranking said portions from highest score to lowest score; and determining location on said sample of said portion with highest score.

2. The method of claim 1 wherein said score depends on said first set of features and said second set of features.

3. The method of claim 1 wherein said first set of features comprises a first array of holes and said second set of features comprises a second array of holes, said second array of holes differing from said first array of holes.

4. The method of claim 1 wherein said test image is from a field of view at a nominal location on said sample based on an external reference coordinate system.

5. The method of claim 1 further comprising storing said test image in a buffer.

6. The method of claim 1 wherein said plurality of portions is from a specified region of interest (ROI) of said test image.

7. The method of claim 1 further comprising discarding said portions with scores below an allowable threshold.

8. The method of claim 4 further comprising comparing location of said portion with the highest score with said nominal location.

9. The method of claim 8 further comprising calculating offsets and scaling factors.

10. The method of claim 4 further comprising capturing a measurement image at a measurement location within said field of view.

11. The method of claim 10 further comprising storing said measurement image in a buffer.

12. The method of claim 10 further comprising acquiring a signal profile of said measurement image.

13. The method of claim 12 further comprising using an edge detection algorithm to measure a critical dimension (CD).

14. The method of claim 1 wherein said score depends on a degree of match between said reference image and said test image.

15. The method of claim 1 further comprising using normalized correlation in said determining of said score for said each of said portions.

16. The method of claim 15 wherein said normalized correlation is not susceptible to linear changes in brightness.